AI

Explaining Humour Style Classifications: An XAI Approach to Understanding Computational Humour Analysis

Humour styles can have either a negative or a positive impact on well-being. Given the importance of these styles to mental health, significant research has been conducted on their automatic identification. However, the automated machine learning models used for this purpose are black boxes, making their prediction decisions opaque. Clarity and transparency are vital in the field of mental health. This paper presents an explainable AI (XAI) framework for understanding humour style classification, building upon previous work in computational humour analysis.

Toward Generating Dynamic Quest Narratives Using Player Behavior

Quests with compelling narratives are notoriously difficult to write and expensive to maintain, yet in most MMORPGs, quests must be churned out constantly to keep the players participating. Quests are often delivered in form of “instance” or parallel universes accessible only to a single player or a small party, and featuring micro-narratives that must necessarily be inconsequential to the world.

Automatic Bill Recommendation for Statehouse Journalists

AI4Reporters is a project designed to produce automated electronic tip sheets for news reporters covering the statehouses (state level legislatures) in the United States. The project aims to capture the most important information that occurred in a bill discussion to allow reporters to quickly decide if they want to pursue a story on the subject. In this paper, we present, discuss and evaluate a module for the tip sheets that is designed to recommend additional bills to investigate for the reporter that receives the tip sheet.

Feature Engineering for US State Legislative Hearings: Stance, Affiliation, Engagement and Absentees

In US State government legislatures, most activity occurs in committees made of lawmakers discussing bills. This paper presents systems to extract legislators' engagement and absence during committee meetings and the stance and affiliation of non-lawmakers making public comments. We propose a system to track the affiliation of organizations in public comments and whether the organizational representative supports or opposes the bill.

Dynamic Procedural Music Generation from NPC Attributes

Procedural Music Generation in Games (PMGG) can enrich the playing experience by providing both entertainment and communication to the player. We present a system that generates unique procedural thematic music for non-player characters (NPC) based on developer-defined attributes and game state. The system responds in real-time to the dynamic relationship between the player and target “boss” NPC. We create a multiplayer 2D adventure game using and evaluate the music generation system by means of user study.

Evaluation of Automatic Text Summarization using Synthetic Facts

In US State government legislatures, most of the activity occurs in committees made up of lawmakers discussing bills. When analyzing, classifying or summarizing these committee proceedings, some important features become broadly interesting. In this paper, we engineer four useful features, two applying to lawmakers (engagement and absence), and two to non-lawmakers (stance and affiliation). We propose a system to automatically track the affiliation of organizations in public comments and whether the organizational representative supports or opposes the bill.

Making California Legislative Process Transparent

This White Paper describes the research and design of the software system called Digital Democracy. The Digital Democracy system will provide its users: individuals with interest in public policy, journalists and media organizations, public interest and government watchdog groups, and lawmakers with direct access to videos and transcripts of State Legislative Committee hearings, as well as analytical tools enabling users of the system to conduct in-depth research of public policy issues and the attitudes of state legislators towards them.

Evolving unsupervised neural networks for Slither.io

Slither. io is a massively multiplayer online game in which up to 500 players control worm-like avatars and consume food to grow with the goal of becoming the largest player while avoiding running into one another. The platform serves as a good testbed for developing AI controlled agents due to its accessibility, mechanical simplicity, and unpredictability. In this paper, we develop a Slither. io bot using neuroevolution of augmenting topologies (NEAT) and compare its performance to that of the best open source bot available online (a high-performing expert system bot).

Composition of basic heuristics for the game 2048

2048 is a simple and intriguing sliding block puzzle game that has been studied for several years. Many complex solvers, often developed using neural nets are available and capable of achieving very high scores. We are, however, interested in using only basic heuristics, the kind that could be conceivably employed by human players without the aid of computation. A common way to implement a 2048 solver involves searching the game tree for the best moves, choosing a move and scoring the game board using some evaluation functions.

Learning alignments from legislative discourse

In this work, we seek to quantify the extent to which a legislator's spoken language indicates their degree of alignment toward an organization that has a taken a documented position on some legislation. To perform this study, we use a corpus of bill discussion transcripts provided by Digital Democracy1. We then apply proven learning methods in the field of natural language processing to predict alignment scores between each member of the California state legislature and a select set of state-recognized organizations.